You are currently reviewing an older revision of this page.
Here is a list of detectors already trained. They can be executed in ContextCapture and ContextCapture Center Master to run Annotation jobs. Each detector was trained:
Meaning, while running on your dataset, each detector type can only be used for the same specific type of job (Annotation job type).
The quality of the detection will depend on the similarity between your dataset and the training dataset’s description.
Please make sure to download detectors matching your ContextCapture version. We recommend you to update to latest In case no detector fits your purpose, you are welcome to submit a help ticket from your personal portal describing your expectations.
Annotation job type
Name & Purpose
Description of training dataset
Illustration
2D Objects
3D Objects
3D Segmentation*
*only as a secondary tool for optimization
Coco
90 classes for everyday life objects: cars, books, chairs, etc…
Update 19
Images: Handheld
Resolution: Not available
Region: multiple
Antennas_v1
1 class for antennas mounted on towers
Images: Drone
Resolution: around 1cm/pix
Region: Multiple
Faces & License plates
1 single class to group faces and license plates and support anonymization workflows
Images: Mobile mapping device - Panoramas
Resolution: N/A
Region: Western Europe
ContextCapture version
2D Segmentation
Segmented mesh
Mesh patches
Cracks
1 class for cracks in concrete infrastructure
Images: Drone + Handheld
Pascal
20 classes for everyday life elements: cars, motorbikes, persons, etc…
3D Segmentation
Buildings A
1 class for buildings in city environment
Pointcloud: RGB - Derived from aerial photogrammetry
Resolution: 20cm
Region: Graz - Western Europe
Ground Occupation
5 classes in city environment: Roofs, vegetation, bridges, power lines, ground
Non-Commercial use
Pointcloud: RGB - Aerial Lidar
Region: Strasbourg - Western Europe
Ground Occupation + 3D objects:
Segmentation-5 classes:Ground, vegetation, power lines, buildings, fences
3D objects: Trucks, cars, poles
Pointcloud: Non colorized - Aerial Lidar
Resolution: 5cm
Region: Dayton - North America
Segmentation-6 classes:Urban furnitures, roofs, facades, trees, shrubs, vertical surfaces
3D objects:Chemineys, vehicles
Pointcloud:RGB - Aerial Lidar
Resolution: 10cm
Region: Hessigheim - Western Europe
Rail
9 classes for usual rail assets: Rails, Signals, PointSensors, etc…
Pointcloud: RGB - Mobile mapping system
Resolution: 3cm
Segmented orthophotos
Roofs – A 1 class for building roofs in city environment
Images: Vertical - aerial mapping camera
Resolution: around 30cm/pix
Roofs – B
1 class for building roofs in city environment
Resolution: around 7.5cm/pix
Geographic area: Christchurch - New Zealand
City - A
6 classes for buildings, High vegetation, low vegetation, vehicles, roads
Resolution: 5cm/pix
Geographic area: Potsdam - Western Europe